Many thanks for posting these links. I am a recent user of Scikit-Learn having
been introduced to it via Kaggle example code. Both tutorials were very helpful
for me, too bad Olivier did not have time to complete all the stuff he planned
to talk about, but I have downloaded both presentations an
Hi Everyone,
In sklearn.cross_validation.cross_val_score, would it make sense to enable
score_func to access the estimator itself, rather than just the true and
predicted labels? My use case would be to use cross_val_score to compute both
classification accuracy and sparsity for each train/test
Hi Michal.
This is implemented in the current development version.
You can now pass "scoring objects" that get the estimator.
See here:
http://scikit-learn.org/dev/modules/model_evaluation.html#score-func-objects
Cheers,
Andy
On 04/06/2013 07:32 PM, Romaniuk, Michal wrote:
> Hi Everyone,
>
> In
Hi Everyone,
In sklearn.cross_validation.cross_val_score, would it make sense to enable
score_func to access the estimator itself, rather than just the true and
predicted labels? My use case would be to use cross_val_score to compute both
classification accuracy and sparsity for each train/test
On Sat, Apr 6, 2013 at 5:00 PM, Andreas Mueller wrote:
> Hi Vladan and welcom to sklearn :)
>
Hi Andy,
thanks for your fast reply and greeting :)
I think what you describe is some particular transductive setting in which
> you have training labels for some classes, but not all.
> Transductive
Hi Vladan and welcom to sklearn :)
I think what you describe is some particular transductive setting in
which you have training labels for some classes, but not all.
Transductive means that you know before-hand which data you want to
predict on (i.e. you can use all your data, have labels on
so
Hello,
I watched both excellent tutorials from PyCon 2013 on YouTube and although
without strong background in statistics, encouraged by this fast food and
Andy's Machine Learning Cheat Sheet on screen, I thought to try something
out.
I have large set of signals with extracted spectral signatures
On Sat, Apr 06, 2013 at 04:07:47PM +0900, Vlad Niculae wrote:
> Now I'll hijack this to talk about the snippet page:
> https://github.com/scikit-learn/scikit-learn/wiki/Useful-Snippets
> It's small, estimator-heavy, unstructured, and I wonder about its
> visibility (the wiki feels a tad internal t
Such a preprocessor would be nice to have in a gist linked from a very
visible list of helpful snippets, but in my opinion it certainly
doesn't belong in the codebase. It breaks TOOWTDY and therefore
chances for bugs to be spotted decrese a lot.
Now I'll hijack this to talk about the snippet page
